Carly Brookfield, DIA CEO, has gone through the latest DVSA Annual Report. Here’s What she had to say about it.
As Mark Twain said… “There are three kinds of lies: lies, damned lies, and statistics.”
The famous quote highlights how numbers and data can be massaged to make a weak argument sound more compelling, or to support any point of view. And, to some degree, that quote could be applied to DVSA’s Annual Report, published this week:
DIA breaks down the main points of the report and shares our insight and insider intel on key elements of the agency’s annual performance review.
Key Takeaways from the 2025-26 Report
- Waiting times are still 3 x the target: 21.5 weeks vs 7-week target. The dial has barely moved and we’ve seen little improvement based on year on year figures, despite 200,000 extra tests delivered. Even by DVSA’s own estimate 450k-500k extra tests are still needed to really shift the dial on waiting times, and another 18 months needed to clear the backlog. And all this 5 years on from lockdown ending and several rounds of measures being announced by Ministers, measures that promised to end this test waiting times misery sooner.
- ADI satisfaction fell, not rose: With a 42% trainer satisfaction score vs a 60% target and down from 47% last year — the one KPI moving backwards while candidate satisfaction improved. The report doesn’t explain the divergence but we know that industry trust and confidence in the regulator is poor after several years of waiting times not limited to just Cat B tests, but also hitting Part 2 and 3 tests and Standards Checks (i.e. stuff that candidates, and the wider public, don’t see and therefore judge the agency on – but we know will ultimately have a negative impact on the candidates and the wider public too).
- “Learner readiness” narrative looks like deflection: DVSA pushes 50% first-attempt fail rate and a readiness campaign, but its own risk register blames examiner capacity and booking abuse — net examiner growth (+188) is nowhere near the scale of the shortfall.
- Fees are going up: frozen for 15+ years, now under active review with Department for Transport and the Treasury, where the focus will be to make fee income cover costs. Put this against an £86.6m net expenditure deficit this year and expect fee rises to land on instructors/pupils.
- ADI/PDI Fee income is up: What the report doesn’t really tell you is that fee income from ADI and PDI licence and test fees have risen by a third in the last 5 years. An FOI request made by DIA last month has revealed that the agency has leapt from a revenue of ££7,667,808.83 in 2021/2022 to a revenue of £10,153,445.55 in 2025/26* This growth is inevitably largely due to the exponential growth in PDI licences and the fact the register now stands at its highest number of registrants since pre-COVID levels. However, what hasn’t increased, in line with revenue growth, is delivery of services to ADIs and PDIs. In fact delivery of ADI and PDI tests have continued to deteriorate with a backlog of Part 2 and 3 tests and only less than 700 Standards Checks being delivered last year, compared to over 12,000 a year when the new style checks were introduced (and immediately pre-COVID)
- ADI pipeline under strain: register is now over 44,000 qualified + 5,500 trainees, 20,000+ new applications, but 1,000+ appeals against refused/revoked trainee licences — all of which results in a rising competition scenario in the market with potential over-supply of trainers against demand, alongside process strain on DVSA to deliver to those applicants
- Systemic/IT risk unresolved: legacy TARS booking system is rated a “very high” risk, but the promised replacement platform is still at Beta stage, with no delivery date given.
- Cancellation costs land on instructors, not just candidates: 6,789 short-notice DVSA cancellations compensated (£654k) in the last year. Compensation goes to candidates only; instructors absorb the disrupted lesson slot.
- No regional data: national 21.5-week average likely masks big centre-by-centre variation — report doesn’t disclose it, why?
And from one of our data crunching, licence fee paying driving instructor members, the following commentary from the coalface
“I have been through DVSA’s 2025–26 Annual Report and Accounts, and there are several areas where the figures do not appear to reconcile, or where DVSA presents activity as success despite the underlying outcome remaining poor.
The strongest points I have identified are:
- DVSA says there was a net increase of approximately 188 driving examiners. Elsewhere, it says the active car examiner workforce reached approximately 1,604 FTE, representing a 7% year-on-year increase. A 7% increase would appear to be closer to 105 FTE, so DVSA needs to explain whether it is mixing headcount, FTE, trainees and operationally productive examiners.
- Test volumes are being used to disguise the failure on waiting times.
DVSA delivered 2.169 million practical tests against a target of 1.975 million, but the national waiting time only reduced from 21.9 weeks to 21.5 weeks. The seven-week waiting-time target is also shown as having “no date set”, which removes any meaningful deadline for delivery.
- The published waiting-time measure may understate the real customer experience.
The measure is based on the earliest week in which at least 10% of appointments remain available and is capped at 24 weeks. This means the headline figure does not necessarily show the median wait, longer waits or how long an individual learner actually waits between first trying to book and taking a test.
- Claims of progress on booking-system integrity sit alongside a very high system risk.
DVSA says it has made significant progress in tackling misuse and improving fairness, but the existing Testing and Registration System is still rated as a “Very High” risk because of security and operational limitations. The ultimate mitigation appears to be replacement of the system rather than evidence that the present controls have resolved the problem.
- My personal favourite! DVSA claims 100% of FOI requests were answered within 20 working days.
The report gives no denominator, methodology, exclusions or breakdown of what counted as a response. It would be useful to obtain the underlying request-handling log and compare the claimed performance with the actual received dates, due dates, response dates, clarifications, extensions and internal reviews.
- Examiner incentive and retention payments are not properly evaluated in the report.
DVSA refers to additional-test awards and exceptional payments, but does not clearly state the total cost, number of genuinely additional tests delivered, cost per additional test or measurable effect on retention and waiting times. - The governance language is stronger than the underlying assurance.
The report refers to the highest standards of governance, transparency and accountability, but the overall internal-audit opinion was only moderate assurance. Knowledge and Information Management and Artificial Intelligence were among the areas receiving limited management assurance, which is particularly relevant to FOI searches, records management and consultation analysis.
My view is that DVSA should be probed more on the examiner-number discrepancy, the waiting-time methodology and the 100% FOI compliance claims. They are all based directly on statements in DVSA’s own signed annual report and should relate to information that already exists.”
What’s your view on the Annual Report and what it tells us about DVSA’s performance?
*Source DIA Freedom of Information Request to DVSA on total annual revenue generated by the agency from ADI and PDI services (covering qualifying test and licence fees, and any other charges for services to ADIs and PDIs).
